IP Australia’s 2018 Annual Statistics – a ‘Damning Report Card’ on Australian Patent Filing?

Last week, IP Australia published its Australian Intellectual Property Report 2018 (‘IP Report’).  This annual report is largely a summary of filing numbers and other statistics, along with additional analysis on selected areas in which IP Australia has ongoing research interests.  Last year’s report, for example, looked at Australia’s performance in collaborations between industry and public research institutes based upon the evidence of jointly-filed patent applications.  This theme is continued in the 2018 report, with a chapter reporting on economic analysis indicating that research grants targeting industry/public research collaborations result in better IP outcomes.  Despite this relatively positive news, with its useful insights and analysis, some media outlets have predictably chosen to focus primarily on raw numbers in order to talk down Australia’s efforts.

For example, InnovationAus reporter Stuart Kennedy has called the 2018 IP Report a ‘damning report card on patent filing’ which (Kennedy says) has caused CSIRO chief Larry Marshall to admit that ‘Australia needs to lift its game’.  Kennedy also wrote (somewhat disparagingly, I felt) last week about the fact that Australia’s top patent applicant for 2017 was ‘poker machine king Aristocrat Technologies’, under the headline ‘a shocking punt on patent filings’.

The numbers in the IP Report do indeed tell us that the number one Australian-based applicant for Australian standard patents in 2017 was Aristocrat Technologies Australia, with 157 filings, followed by CSIRO (45), the University of Queensland (18), Bluescope Steel (15) and Monash University (also 15).  But so what?  Aristocrat is a commercial operation, with its own business justifications for seeking patent protection for its innovations, which are no doubt driven by the market in which it operates and its ability to secure a competitive advantage from its patents.  In this context, each individual patent filing may represent a relatively minor innovation, in technical terms, while still making a commercially useful contribution to the overall portfolio. 

Research organisations, such as CSIRO and universities, on the other hand, are expected to invest in substantial and longer-term scientific and technological advances, often at a pre-commercial stage of development.  We would hardly praise them for squandering precious funds on hundreds of individually trivial patent applications, merely for the sake of making their filing numbers look better.

It has to be said, however, that IP Australia does not exactly help the situation in the way it has presented some of the numbers.  It would be very easy (though wrong) to take the following messages away from the patents chapter of the 2018 IP Report (these are all direct quotes, and each appears in the first sentence of a paragraph in the report):

1. ‘While applications grew overall in 2017, applications for standard patents by Australian residents decreased by about five per cent’;
2. ‘Non-resident filings increased by two per cent in 2017 to 26 403, which was 91 per cent of all filings’;
3. ‘World patent filings have been growing strongly since 2010, averaging around eight per cent annual growth to 2016, while Australia's growth has averaged about three per cent over the same period’;
4. ‘Patent grants to Australian residents in 2017 fell by 17 per cent compared to 2016 and now make up just five per cent of the total.’

If this were all there is to the numbers, the description ‘damning report card’ might indeed be apropos.  Fortunately, the situation is not nearly as bad as these figures suggest.

Is the ‘Report Card’ Really So ‘Damning’?

The quotations listed above present a dismal picture of Australian patenting performance.  And while further nuances become apparent upon a more careful reading of the report, IP Australia Director General, Patricia Kelly, should hardly be surprised to find herself compelled to explain these numbers, as she did to Stuart Kennedy in his ‘shocking punt’ article:

People get very fixated on the figures whereby Australians are only filing between six and eight percent of total patents in Australia.

That’s not such a bad thing because we produce about three percent of the world’s research output so you would expect that the vast majority of knowledge is created overseas so you would expect the vast majority of patents to come from overseas.

There is much more to the Australian patent (and other IP rights) data than meets the eye in IP Australia’s brief overview, as all interested stakeholders will be able to confirm for themselves once the 2018 Intellectual Property Government Open Data (IPGOD) data set is released.  The IP Report indicates that its publication is accompanied by the release of IPGOD 2018, however this is not quite accurate, and the new data release is still in the final stages of preparation at the time of writing.  This year I have been engaged by IP Australia to assist in processing and preparation of the data, which means that I have had access to a portion of the raw data on which IPGOD is based.  While it would be inappropriate for me to pre-empt the formal release of IPGOD 2018, which will enable far more detailed analysis, here are a few additional facts about Australians’ patent filing performance to be thinking about in the meantime.

1. Although innovation patent filings fell quite steeply in 2017, Australian resident filings remained substantially steady (from 1,087 in 2016 to 1,057 in 2017, which IP Australia chose to characterise as a ‘three per cent decline’).  More significantly, Chinese abuse of the innovation patent system dropped sharply, with Chinese-originating filings falling by more than half, from 879 in 2016 to 418 in 2017.  This positive development explains most of the overall decline in innovation patent applications.
2. While US-based applicants remain the largest users of the Australian patent system, and Australian residents did show a small decline in 2017, Australians remain comfortably the second largest users of their own national patent system, filing around 1000 applications more in 2017 than third-placed Japan.
3. In 2015 and 2016 (the most recent years for which I have reliable data) Australian residents were ranked 16^th among US patent applicants, which is arguably not too bad for a country ranked 53^rd in the world by population.  Indeed, the number of US patent applications naming at least one Australian-resident inventor in 2015 (2341) was directly comparable to the number of Australian applications filed by Australian residents in that same year (2330).
4. It is true that commercial organisations dominate the filing of standard patent applications in Australia (whether they be resident companies, like Aristocrat, or foreign entities such as top-three-placed Halliburton Energy Services, Qualcomm and Samsung), however the picture is quite different if we look at provisional filings.  Here, the top six applicants are research/tertiary institutions – CSIRO, Monash University, the University of NSW (via NewSouth Innovations Pty Ltd), and the Universities of Sydney, Melbourne and Queensland – which collectively filed 223 provisional applications in 2017.  Provisional filings are often a better indication of innovative – and potentially commercial – research output. Notably, these inventions are typically ‘pruned’ at the 12-month (PCT filing) and 30-month (national filing) stages, as commercial opportunities are progressively re-evaluated, which should be regarded as a positive and prudent strategy by our research organisations.

Conclusion – Raw Data Needs More Analysis

Of course there is certainly much room for improvement in Australia’s performance in innovation and IP capture/protection, as the Australian government body Innovation and Science Australia (ISA) amply identified via the 30 recommendations in its Australia 2030 strategy.  However, using raw (or, at best, undercooked) patent filing statistics as a stick with which to beat Australian industry and research organisations is a pointless, and even counter-productive, exercise, which is liable to create a false impression of their performance.

There is no reason to suppose, for example, that Aristocrat's business reasons for filing relatively large numbers of Australian patent applications tells us anything about the value of any individual application (or how many of those applications will actually result in granted patent rights).  It certainly does not say anything about the social value of gaming/gambling technology relative to any other fields of endeavour!  Nor is CSIRO's 45 Australian applications at all insignificant.  Over a 20-year maximum patent term, maintaining this rate of filing represents a potential pipeline on the order of 900 patented inventions, and it is likely that corresponding applications have been filed in a number of other jurisdictions.  There is no question that CSIRO's global patent portfolio represents a very substantial financial investment, of which the Australian filings alone represent only one component.  And who knows – somewhere in 2017's 45 patent applications there may be another wireless LAN!

I look forward to presenting further analysis of the data once the IPGOD 2018 data set is formally released.

Tags: Australia, Innovation policy, Patent analytics